Breaking an electronically preferred symmetry by steric effects in a series of [Ir(biph)X(QR3)2] compounds (X = Cl or I, Q = P or As)

Article abstract of DOI:10.1039/a804840a

A hybrid quantum mechanical/molecular mechanics IMOMM (B3LYP : MM3) method has been applied to a series of five-coordinate 16-electron d⁶ ML₅ Ir^III compounds having a relatively flat potential for a distortion from Y to T geometry and for which crystal structures have been obtained. In this series of type [Ir(biph)X(QPh₃)₂] (biph = biphenyl-2.2′-diyl; Q = P, X = Cl, 2a; Q = As, X = Cl, 2b; Q = P, X = I, 2c), the halide is found to lie in the (biph)Ir plane but off the C₂ axis of the {Ir(biph)Q₂} fragment by a variable angular distortion φ. While φ = 0 is preferred electronically for [Ir(C₄H₄)Cl(PH₃)₂], the steric bulk of the real systems 2a-2c leads to φ taking experimental values of 8.2-17.2°. The observed deviation of the halide from the C₂ axis is shown by IMOMM to be the result of a direct interaction of the phenyl substituents of the axial ligands with the equatorial ligands and not to an electronic effect. The crystal structures for 2b and 2c have been determined.

Full text of DOI:10.1039/a804840a

View Article Online / Journal Homepage / Table of Contents for this issue
Breaking an electronically preferred symmetry by steric e†ects in a
series of [Ir(biph)X(QR ) ] compounds (X = Cl or I, Q = P or As)
3 2
Gregori Ujaque,a Feliu Maseras,*,¤,a Odile Eisenstein,*,a Louise Liable-Sands,b
Arnold L. Rheingold,*,b Wenbin Yaoc and Robert H. Crabtree*,c
a L SDSMS (UMR 5636) Case Courrier 14, Universite de Montpellier 2, 34095, Montpellier,
Cedex 5, France
b Department of Chemistry, University of Delaware, Newark, DE, 19716, USA
c Department of Chemistry, Y ale University, New Haven CT , 06520-8107, USA
A hybrid quantum mechanical/molecular mechanics IMOMM (B3LYP : MM3) method has been applied to a
series of Ðve-coordinate 16-electron d6 ML IrIII compounds having a relatively Ñat potential for a distortion
5
from Y to T geometry and for which crystal structures have been obtained. In this series of type
[Ir(biph)X(QPh ) ] (biph \ biphenyl-2.2@-diyl; Q \ P, X \ Cl, 2a; Q \ As, X \ Cl, 2b; Q \ P, X \ I, 2c), the
3 2
halide is found to lie in the (biph)Ir plane but o† the C axis of the MIr(biph)Q N fragment by a variable angular
2
2
distortion z. While z\ 0 is preferred electronically for [Ir(C H )Cl(PH ) ], the steric bulk of the real systems
4
4
3 2
2aÈ2c leads to z taking experimental values of 8.2È17.2¡. The observed deviation of the halide from the C axis
2
is shown by IMOMM to be the result of a direct interaction of the phenyl substituents of the axial ligands with
the equatorial ligands and not to an electronic e†ect. The crystal structures for 2b and 2c have been determined.
Quantum chemical calculations have attained an impressive
level of reliability in recent years and quantitative studies are
possible even on reasonably large molecules. To save compu-
tational time, it is nevertheless still usually necessary to sim-
plify the ligand set by replacing PR by PH , for example.
This means that when there is a signiÐcant deviation between
the quantum chemical and experimental results on a given
system, one cannot immediately tell whether the deviation is
missing vertex and the p-donor ligand is located in the basal
site.
3
3
due to the steric e†ects of the real PR ligand or to some
3
nonsteric electronic e†ect that has not been well represented
in the calculations.
Hybrid
quantum
mechanicsÈmolecular
mechanics
(QM/MM) methods have recently become available that can
help resolve such ambiguities. In these, the core of the mol-
ecule, consisting of the metal and the immediate ligand sphere,
is represented by quantum chemical methods, and the exterior
part, comprising the substituents on the ligands, is represented
by molecular mechanics. In this way a large molecule can be
treated realistically without requiring prohibitive amounts of
computer time. Here, we report application of a combined
(QM/MM) hybrid method IMOMM (B3LYP : MM3) to the
title complexes. This particular hybrid method1 has already
proved successful in computing steric e†ects in several
transition-metal complexes.2
Two experimental examples, discovered almost simulta-
neously in the groups of Crabtree and of Caulton,4h8 showed
that in certain circumstances it is possible to break the sym-
metry in what would otherwise be a pure Y-type molecule. In
spite of having two identical ligands in the equatorial plane
these compounds adopt structure 1c intermediate between Y
and T geometries, denoted Y/T. For [IrH Cl(PMButN Ph) ]
Following the original experimental discovery,3a various
theoretical studies have clearly established that Ðve-
coordinate d6 16-electron complexes with one equatorial
p-donor ligand (labeled L@ in diagram 1a), show a preference
eq
2
2
2
for a Y structure, 1a, This resembles a trigonal bipyramidal
(1d) the Y/T structure is found in the solid state by neutron
di†raction4 and is maintained in solution as shown by 1H
NMR studies at low temperature.5 IMOMM (B3LYP : MM3)
calculations revealed that the observed deviation from Y
geometry is a result of steric interactions between the phos-
phine substituents and the equatorial ligands.6 The potential
energy surface linking Y and T geometries is known to be
Ñat3d and it is no doubt this factor that permits the existence
of structures with a geometry intermediate between the two
pure forms.
structure but has one small angle (\90¡) between two equato-
rial ligands.3bhd In such a Y structure, the MwL@ bond
eq
deÐnes the mirror plane which necessarily bisects the angle
between the two identical equatorial ligands. A T structure 1b
is not very di†erent in energy, however, and in certain cases
can be the ground state. This structure is octahedral with one
* E-mail: odile.eisenstein=lsd.univ-montp2.fr; arnrhein=udel.edu;
robert.crabtree=yale.edu
¤ Present address: Unitat de Qu•mica F•sica, EdiÐci C.n., Universitat
Autonoma de Barcelona, 08193, Bellaterra, Catalonia, Spain;
feliu=klingon.uab.es
This paper probes the generality of these distortions and
reports a series of 16-electron IrIII complexes where the IrH
2
New J. Chem., 1998, Pages 1493È1498
1493

View Article Online
LiI
[Ir(biph)Cl(PPh ) ] ÈÈÈ [Ir(biph)I(PPh ) ]
3 2 3 2
(3)
acetone
The Ðve-coordinate, 16-electron species 2aÈ2c are not readily
converted to the corresponding six-coordinate species
[Ir(biph)X(QPh ) ]. In more recent work9 we Ðnd that the
3 3
reaction of [Ir(biph)Cl(cod)] with SbPh gives
a six-
3
coordinate product, [Ir(biph)Cl(SbPh ) ], presumably because
3 3
SbPh is a slightly less sterically bulky ligand largely as a
3
group of the above system is replaced by an iridium diaryl.
This new system shows the same types of distortions and pro-
vides a good quantitative test of the hybrid methods men-
tioned above. The new series of complexes has the advantage
that only non-hydride ligands are present, permitting deÐni-
tive structural study without recourse to neutron di†raction.
result of its longer IrwSb bond. The sterically unencumbered
PMe also gives a six-coordinate [Ir(biph)Cl(PMe ) ].9 This
3
3 3
suggests that the formation of the Ðve-coordinate compounds
2aÈ2c is at least in part the result of steric e†ects, although the
presence of the high trans-inÑuence biph ligand no doubt also
encourages Ðve-coordination.
Crystals suitable for crystallographic study were grown
from CH Cl solution by slow evaporation. The crystallo-
Results and Discussion
2
2
We showed7 in 1993 that biphenylene undergoes a CwC
graphic work (Table 1) gave the distance and angular data of
Table 2. The corresponding ORTEP diagrams are shown in
Fig. 1 and 2. The crystal of compound 2b contained two inde-
pendent molecules having very similar but not identical metric
parameters.
bond activation reaction with [Ir(cod)Cl] (cod \ 1,5-cyclo-
2
octadiene) to give
a
biphenyl-1,2-diyl (\biph) complex,
[Ir(cod)(biph)Cl] , where the biph ligand is the organometallic
2
analogue of 2,2@-bipyridine [eqn. (1)]. The complex was found
to react readily [eqn. (1)] with PPh to give a 16-electron Ðve-
Both 2b and 2c have similar structures to that of
3
coordinate IrIII species, [Ir(biph)Cl(PPh ) ] (2a) which was
structurally characterized. Five-coordinate IrIII species like 2
[Ir(biph)Cl(PPh ) ] (2a) which was determined previously.8 In
3 2
3 2
all cases, the (biph)IrQ fragment has near C symmetry but
2
2v
the halide, while remaining in the equatorial biph plane (the
sum of the three equatorial LwIrwL angles is 359.87È360¡),
lies signiÐcantly o† the C axis so that one biph carbon,
2
denoted C
, becomes more trans than cis, and the other,
trans
denoted C , becomes more cis than trans to the halide group.
Fig. 3 illustrates the situation by showing the equatorial
cis
ligands. All the molecules of type 2 studied to date therefore
have essentially the same geometry, lying between the ideal Y
and T forms, but they show di†erent deviations from pure Y
geometry. We have used the angles x , x and z shown in
1
2
Fig. 3 as a measure of this deviation in the following dis-
cussion. The fact that C , C , Ir and X are always in the
cis trans
same plane allows the value of z to be derived from x and x
by eqn. 4.
1
2
are expected to have a pure Y structure owing to the presence
of only one equatorial p-donor ligand, a halide.
At the time of its discovery, it was surprising to Ðnd that in
2a the experimental structure was neither Y nor T but inter-
mediate between the two situations, as shown in diagram 3c.
This was initially thought to be an electronic e†ect but work
z[ o 1(x ] x ) [ x o
(4)
2
1
2
1
A pure Y geometry should correspond to a z value of zero
and a pure T geometry should lead to a z value close to 45¡.
The values found in the present work lie between 8.2 and 17.2¡
and therefore indicate a geometry closer to Y than T.
Table 1 X-Ray crystallographic data for 2b and 2ca
Compound
2b
2c
Formula
Space group
a/Ó
C
P2 /c
H
As ClIr
C
P2 /n
14.5086(2)
10.92350(8)
25.52790(10)
103.6310(10)
3931.83(6)
995.82
H P IIr
48 38
2
48 38 2
1
1
23.5757(3)
19.10520(10)
19.1809(2)
113.2260(10)
7939.30(19)
992.27
b/Ó
to be described below now makes clear that we are seeing a
steric e†ect. As a test, we wanted to compare a series of com-
plexes of type 2 to see if the deviation from the ideal Y
geometry changes with the steric e†ects of the ligands.
c/Ó
b/¡
V /Ó3
MW
q(calc)/g cm~3
1.660
8
1.42È28.17
14939
1.682
4
1.48È28.26
6706
Z
Synthesis and structural study
h Range/¡
No. of data used
No. of parameters
reÐned
In the present work, we have synthesized and structurally
937
467
characterized two related species of type 2 with ligands of
slightly di†erent steric size: [Ir(biph)Cl(AsPh ) ] (2b) and
3 2
[Ir(biph)I(PPh ) ] (2c). These have now been prepared in
3 2
R[I [ 2r(I)]b
3.28%
8.22%
3.15%
7.87%
R [I [ 2r(I)]b
w
good yield as air-stable orange (2b) or red (2c) crystals by the
a Details in common: T \ 173(2) K, scan type xÈ2h, k \ 0.71069 Ó,
largest shift/error \ 0.2, empirical DIFABS absorption correction.
bFunction minimized: R \ &pF o [ o F p/o F o and R \ [&w(o F o
standard routes shown in eqns. 2 and 3.
AsPh3
o
c
o
w
o
[ o F o)2/&wF2]1@2, where w \ 4F2/r2(F )2.
[Ir(cod)Cl(biph)] ÈÈÈ [Ir(biph)Cl(AsPh ) ]
(2)
c
o
o
o
2
3 2
1494
New J. Chem., 1998, Pages 1493È1498

View Article Online
Table 2 Selected bond distances (Ó) and angles (¡) for 2b and 2c; X \ halide, Q \ P or As
2b (1st molecule)a
2b (2nd molecule)a
2c
IrwX
2.3950(10)
2.4359(4)
2.4362(4)
2.041(4)
2.3967(10)
2.4350(4)
2.4332(4)
2.050(4)
2.7197(3)
2.3374(12)
2.3509(12)
2.051(4)
IrwQ(1)
IrwQ(2)
IrwC
b
trans
IrwC
b
2.035(4)
2.034(4)
2.024(5)
cis
XwIrwQ(1)
XwIrwQ(2)
92.99(3)
90.37(3)
149.27(12)
132.93(13)
173.28(2)
77.8(2)
88.83(3)
93.35(3)
151.21(12)
130.96(14)
174.04(2)
77.7(2)
91.92(3)
91.77(3)
158.1(2)
123.75(14)
172.62(5)
78.1(2)
XwIrwC
trans
XwIrwC
cis
Q(1)wIrwQ(2)
C
z
wIrwC
trans
cis
8.2(4)
10.1(4)
17.2(4)
a The crystal of 2b contained two independent molecules; the atom numbering of the Ðrst has been given unmodiÐed [e.g. C(1)] and the second in
primes [e.g. C(1@)]. b The carbon atoms of the biph bound directly to Ir are labeled as being cis or trans with reference to the halide because of the
slightly di†erent numbering system in the two compounds. C is C(1) in the Ðrst molecule of 2b, C(12@) in the second molecule of 2b and C(7) in
cis
2c; similarly, C
is C(12) in the Ðrst molecule of 2b, C(1@) in the second molecule of 2b and C(1) in 2c.
trans
We Ðnd that both the ligand conformations and the z
values in the two molecules of 2b are essentially the same,
suggesting that packing forces have only a minor role in deter-
mining the distortion. This is a rare case in which an
organometallic molecule with a relatively Ñat bending poten-
tial (see below) has been studied in two independent forms.
The low inÑuence of packing forces on the distortion sug-
gested by the present work may not be general, however,
because ionic solids in particular might be expected to have a
much higher e†ective internal pressure and could therefore
show a greater structural variability depending on the details
of the packing.
C symmetry which is present in the calculated structure and
lacking in the experimental system. The computed geometry
2
had C symmetry despite the fact that the geometry opti-
2
mization was carried out with no symmetry restrictions and
Theoretical study
Previously reported calculations8 at the extended Huckel
theory level seemed to indicate an electronic origin for the
deviation, z. The structure of the simpliÐed system [Ir-
(C H )Cl(PH ) ] was thus optimized at the B3LYP level. The
4
4
3 2
results show that the geometry of the complex is close to that
of the full biph system with the exception of the presence of a
Fig. 2 ORTEP diagram of 2c. Carbons C(72)ÈC(74) are disordered
over two positions and were reÐned isotropically. Thermal ellipsoids
are plotted at 50% probability and the hydrogen atoms are omitted
for clarity
Fig. 3 Ligands in the equatorial plane of complex 2, showing the
deviation of the halide from the ideal C axis (dotted) of a pure Y
2
structure. QR ligands are located above and below the equatorial
3
Fig. 1 ORTEP diagram of 2b. Only one of the two independent but
chemically equivalent molecules is shown. The other has essentially
the same conformation. The thermal ellipsoids are plotted at 50%
probability and the hydrogen atoms are omitted for clarity
plane. Angle z is the deviation of the X group from the C axis, x is
2
1
the C wIrwX angle and x is the C wIrwX angle. X always lies
cis
in the C , C
cis trans
2
trans
, Ir plane
New J. Chem., 1998, Pages 1493È1498
1495

View Article Online
started from a non-symmetrical structure. The IrwC distances
are reasonably well represented (calcd., 2.06 Ó; exptl. average,
2.00 Ó), the alternating long and short distances within the
metallacycle are also well reproduced (calcd., 1.355, 1.459 Ó;
exptl. 1.415, 1.440 Ó). The signiÐcant di†erence between
theory and experiment for the CxC of the metallacycle is
clearly due to it being part of a phenyl ring in the biph
complex, in which case the double bond is delocalized. The
calculated IrwCl distance of 2.449 Ó is reasonably close to the
experimental value of 2.380 Ó as is also the case for the IrwP
distances (calcd., 2.331 Ó; exptl., 2.345 Ó). As in the experimen-
tal system, the metallacycle is planar. A pure Y structure for
[Ir(C H )Cl(PH ) ] agrees with the expectations based on the
the axis being very Ñat, good agreement with experiment was
obtained.
On going from PH to PPh (X \ Cl), all the calculated
3
3
metalÈligand distances increase slightly, probably as a conse-
quence of the increased steric bulk around the metal. The
CwC bond of the metallacyclopentadiene is not changed on
moving from the IrC H to the Ir(biph) model since the adja-
4
4
cent phenyl rings of biph are represented only at the MM3
level and therefore delocalization of the electrons of the CxC
bond is not possible.
The calculated value of z increases on going from Cl to I
(L \ PPh ) as is also the case for the experimental results
3
(calc., 14.6; exptl., 17.2¡). The increased distortion is clearly
4
4
3 2
presence of one equatorial p donor (Cl).3 Our previous
proposal8 that the geometrical constraint of the biph ligand
helps cause the distortion has therefore been discarded.
associated with a bigger steric interaction with the larger
halide. Changing Q from P to As (X \ Cl) diminishes the
experimental distortion by only one degree [z\ 9.15¡(av.)];
the calculations show no change (calc., 11.3¡). The longer
IrwQ distance presumably causes the ligand cone angle to
decrease.
Having established that 2 has an electronic preference for a
Y structure with z\ 0¡, the observed deviation in 2aÈ2c
seemed likely to be a steric e†ect. This makes the series of
compounds of type 2 suitable for study by the hybrid com-
bined IMOMM (B3LYP:MM3) method. Examination of the
experimental structures of 2aÈ2c showed that in each case the
same ligand conformation is adopted. This conformation was
therefore taken as the departure point for the IMOMM study
so as to minimize the potential problem of falling into a series
of local minima of no relevance to the experimental solid-state
conformation. The minimization was carried out with the
object of seeing how well the z value that emerged in each
case matched the experimental value and if the trend of the
values matched the experimental trend.
The calculations thus closely mimic the experimental
system, even though the substituents on QR are represented
3
in a purely MM (steric) manner; this suggests that the inter-
actions between the substituents and the other ligands are
essentially controlled by steric factors. Previous work in this
area2b,9 has shown that steric e†ects in organometallic com-
pounds tend to be dispersed over numerous centers and no
particular atom or group can be considered as the dominant
contributor. Compound 2 has even more atoms than the com-
plexes studied previously and so the possibilities for dispersal
of steric e†ects are even greater. For this reason, we did not
conduct any analysis to attempt to estimate the relative size of
the interatomic repulsions. It is also very satisfying that the
IMOMM hybrid method is able to reproduce subtle structur-
al changes from Y to T even on a potential energy surface
known from ab initio calculations to be very Ñat.3bhd
The IMOMM calculations on [Ir(biph)X(QR ) ] always
3 2
reproduce the experimental conformation, indicating that this
is a local minimum for the isolated complex and that the
crystal packing is not a determining factor. The calculations
also show a displacement of the halide from its electronically
preferred position on the C axis in a direction which allows
2
Experimental
Syntheses
the IrwX bond to stay coplanar with the metallacycle.
However, as in the case of the IMOMM calculation on
[IrH ClL ], the geometrical results are not quantitatively
General procedures and materials. All manipulations were
performed under a dry nitrogen atmosphere using standard
Schlenk techniques. Solvents were dried by standard pro-
cedures. All the reagents were used as received without further
puriÐcation. [Ir(cod)Cl(biph)] and [Ir(biph)Cl(PPh ) ] were
2
2
satisfactory if the radius of Cl is maintained at its standard
value contained in the MM3 program because this is the value
appropriate for organic chlorides. With the standard radius,
[Ir(biph)Cl(PPh ) ] shows only a slight distortion away from
3 2
2
3 2
C
symmetry (z\ 1.4¡). If the Cl radius is increased to
prepared by literature methods.7
2v
account for its ionic nature using the procedure described in a
previous paper,6 the optimized structure now comes remark-
ably close to the experimental one (calc. z\ 11.4; exptl.,
10.15¡). For all Q and X studied, the displacement, as mea-
sured by x , x and z, is given in Table 3. The same distorted
1H and 31P NMR spectra were recorded on a GE Omega
300 spectrometer. Elemental microanalyses were performed by
Robertson Microlit Laboratories.
Chloro(biphenyl-2,2º-diyl)bis(triphenylarsine)iridium(III)(2b).
1
2
geometry was obtained from calculations starting both from
the experimental non-symmetrical arrangement and starting
A
suspension of [Ir(cod)Cl(biph)] (0.20 g, 0.21 mmol)
2
and triphenylarsine (0.28 g, 0.92 mmol) in CH Cl (10 mL)
2
2
from a symmetrical C arrangement. The latter arrangement
was stirred at room temperature for 5 h under N . The
2v
2
was tested speciÐcally in the case of [Ir(biph)Cl(PPh ) ] to
resulting red-orange product was Ðltered o†, washed with
3 3
ensure that no other local minimum is present. In spite of the
Et O (2 ] 10 mL), and dried in vacuo. Yield: 0.28 g (0.28
2
potential energy surface for the displacement of the Cl from
mmol, 67%). 1H NMR (CD Cl , 298 K): d 7.49 (2H, d,
2
2
Table 3 Results of the IMOMM and (B3LYP : MM3) calculation on [Ir(biph)XL ]
2
Calc.
Exptl.
Compound
L
X
x
x
z
x
x
z
1
2
1
2
2aa
2b
PPh
AsPh
AsPh
PPh
PH
3
Cl
Cl
Cl
I
130.6
130.7
130.7
128.0
142
153.4
153.3
153.3
155.3
142
11.4
11.3
11.3
14.6
0
131.3
132.93
130.96
123.75
È
151.4
149.27
151.21
158.1
È
10.15
8.2
10.1
17.2
È
3
3
3
2bºb
2c
3
2dc
Cl
1
Angles in degrees. x \ C wIrwX; x \ C wIrwX; z\ o (x ] x ) [ x o. a With the standard MM3 value of the Cl radius (adapted for
1
cis
2
trans
2
1
2
1
organic chlorides) x and x are 140.4 and 143.2¡ (z\ 1.4¡). b Data for the second molecule in unit cell. c Theoretical value only; compound not
1
2
synthesized.
1496
New J. Chem., 1998, Pages 1493È1498

View Article Online
J \ 7.2 Hz), 7.2È7.4 (36H, br, Ph), 6.47 (2H, t, J \ 7.2 Hz,
biph), 6.33 (2H, t, J \ 7.2 Hz, biph), 6.31 (2H, d, J \ 7.2 Hz,
from the values assigned to P-C-C and P-C-C-R. Torsional
contributions involving dihedral angles with the metal atom
in terminal position were set at zero. The values of the radii of
Cl and I in the MM3 program were modiÐed to take into
account their greater negative charge when bonded to a
transition-metal center according to the procedure described
previously:6 radii for Cl and I were 2.47 Ó and 2.71 Ó, respec-
tively. All geometrical parameters were optimized without
symmetry restrictions except for the bond distance between
the QM and MM regions of the molecules. These were frozen
at 1.420 (PÈH). 1.532 (AsÈH), 1.101 Ó (CÈH) in the QM part;
and 1.828, 1.943 (AsÈC), and 1.434 Ó (CÈC) in the MM part.
The starting point of all geometry optimizations was the
crystal structure coordinates for the iridium complexes.
biph). Anal. Calcd. (found) for C
(57.88); H, 3.86 (4.00%).
H
As ClIr: C, 58.10
48 38
2
Iodo(biphenyl - 2,2º - diyl) bis (triphenylphosphine) iridium (III)
(2c). A mixture of [Ir(biph)Cl(PPh ) ] (0.10 g, 0.11 mmol) and
3 2
LiI (0.14 g, 1.0 mmol) in acetone (20 mL) was stirred at
ambient temperature for 12 h, during which time the color
changed from red-orange to red. The solvent was removed
under reduced pressure, and CH Cl (20 mL) was added to
2
2
dissolve the red precipitate. The solution was Ðltered through
Clite, the Ðltrate was reduced to 5 mL in vacuo, followed by
addition of Et O (15 mL) to precipitate a red product, which
2
was Ðltered o†, washed with Et O (2 ] 10 mL), and dried in
2
vacuo. Yield: 78 mg (0.078 mmol, 71%). 1H NMR (CD Cl ,
2
2
298 K): d 7.2È7.4 (38H, br, Ph and biph), 6.46 (2H, t, J \ 7.3
Hz, biph), 6.34 (2H, dt, J \ 1.1, 7.3 Hz, biph), 6.21 (2H, dd,
Acknowledgements
J \ 1.1, 7.3 Hz, biph). Anal. Calcd. (found) for C
H
P IIr:
We thank NSF for support (R.H.C.) and for their support
(A.L.R.) of the purchase of a CCD-based di†ractometer. We
thank the CNRS for a research associate position (F.M.) and
the UAB for supporting G.U.Ïs stay in Montpellier.
48 38 2
C, 57.90 (58.08); H, 3.84 (3.80%).
Crystallographic structural determination
Crystal, data collection and reÐnement parameters are given
in Table 1. Data was collected on a Siemens P4/CCD di†rac-
tometer. The systematic absences in the di†raction data are
uniquely consistent with the space groups reported. The struc-
tures were solved by direct methods, completed by subsequent
di†erence Fourier synthesis and reÐned by full-matrix least-
squares procedures. An empirical absorption correction was
applied, based on a Fourier series in the polar angles of the
incident and di†racted beam paths and was used to model an
absorption surface for the di†erence between the observed and
calculated structure factors.10 The asymmetric unit of 2b con-
sists of two independent molecules. Three carbon atoms of
both of the biph rings of 2c, equally disordered over two posi-
tions, were reÐned isotropically, and the corresponding hydro-
gen atoms were ignored owing to this disorder. All other
non-hydrogen atoms were reÐned with anisotropic displace-
ment coefficients and all other hydrogen atoms were treated
as idealized contributions. Five of the remaining peaks in the
Ðnal di†erence map of 2b (1.06 to 1.54 e Ó~3) were in chemi-
cally unreasonable positions 1.18È1.44 Ó from the Ir and were
considered as noise.
References
1 F. Maseras and K. Morokuma, J. Comput. Chem. 1995, 16, 1170.
2 (a) T. Matsubara, F. Maseras, N. Koga and K. Morokuma, J.
Phys. Chem. 1996, 100, 2573; (b) Y. Wakatsuki, N. Koga, H.
Werner and K. Morokuma, J. Am. Chem. Soc., 1997, 119, 360; (c)
G. Ujaque, A. C. Cooper, F. Maseras, O. Eisenstein and K. G.
Caulton, J. Am. Chem. Soc., 1998, 120, 361; (d) F. Maseras and O.
Eisenstein, New. J. Chem., 1998, 22, 5.
3 (a) H. Werner, A. Hohn and M. Dziallas, Angew. Chem., Int. Ed.
Engl., 1986, 25, 1090; (b) Y. Jean and O. Eisenstein, Polyhedron,
1988, 7, 405; (c) I. E.-I. Rachidi, O. Eisenstein and Y. Jean, New J.
Chem., 1990, 14, 671; (d) J.-F. Riehl, Y. Jean, O. Eisenstein and M.
Pelissier, Organometallics, 1992, 11, 729.
4 A. Albinati, V. I. Bakhmutov, K. G. Caulton, E. Clot, J. Eckert, O.
Eisenstein, D. G. Gusev, V. V. Grushin, B. E. Hauger, W. Kloos-
ter, T. F. Koetzle, R. K. McMullan, T. J. OÏLoughlin, M. Pelissier,
J. S. Ricci, M. P. Sigalas and A. B. Vymenits, J. Am. Chem. Soc.,
1993, 115, 7300.
5 B. E. Hauger, D. G. Gusev and K. G. Caulton, J. Am. Chem. Soc.,
1994, 116, 208.
6 G. Ujaque, F. Maseras and O. Eisenstein, T heor. Chem. Acc., 1997,
96, 146.
7 Z. Lu, C.-H. Jun, S. de Gala, M. Sigalas, O. Eisenstein and R. H.
Crabtree, J. Chem. Soc., Chem. Commun., 1993, 1877.
8 Z. Lu, C.-H. Jun, S. de Gala, M. Sigalas, O. Eisenstein and R. H.
Crabtree, Organometallics, 1995, 14, 1168.
All software and sources of the scattering factors are con-
tained in the SHELXTL (5.3) program library.11
CCDC reference number 440/062.
Computational details
9 G. Barea, F. Maseras, Y. Jean and A. Lledo
35, 6401.
10 N. Walker and D. Stuart, Acta Crystallogr., Sect. A, 1983, 39, 158.
11 G. Sheldrick, Siemens XRD, Madison, WI, 1996.
` s, Inorg. Chem., 1996,
Pure quantum mechanical calculations on the model systems
[Ir(C H )X(QH ) ] were carried out with Gaussian 94.12
4
4
3 2
Pseudo potentials were used for representing the 60-electron
core of Ir,13 the 10-electron core of P and Cl, the 18-electron
core of As and the 26-electron core of I.13b
12 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Peterson,
J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G.
Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefa-
nov, A. Nanyakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W.
Chen, M. W. Wong, J. L. Andres, G. S. Replogle, R. Gomperts, R.
L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P.
Stewart, M. Head-Gordon, C. Gonzalez and J. A. Pople, Gaussian
94: Revision, D. I. Gaussian, Inc., Pittsburgh, PA, 1995.
13 (a) P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299 (b) P.
J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 284.
14 M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S.
Gordon, D. J. Defrees and J. A. Pople, J. Chem. Phys., 1982, 77,
3654.
15 A. Hollwarth, M. Bohme, S. Dapprich, A. W. Ehlers, A. Gobbi, V.
Jonas, K. F. Kohler, R. Stegmann, A. Veldkamp and G. Frenking,
Chem. Phys. L ett., 1993, 208, 237.
16 W. J. Hehre, R. DitchÐeld and J. A. Pople, J. Chem. Phys., 1972,
56, 2257.
17 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5468; (b) C. Lee, W.
Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785; (c) P. J. Step-
hens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, J. Phys.
Chem., 1994, 98, 11623.
The associated double-f basis set with a LANL2DZ
contraction12 was used for the Ir, P, As, Cl and I atoms. A
polarization d shell was added for the P,14 Cl,14 I15 and As15
atoms. The C and H atoms had a valence double-f basis set.16
Full geometry optimizations were carried out at the B3LYP
level. 17
IMOMM
calculations
were
performed
on
[Ir(biph)X(QPh ) ] (X \ Cl, I; Q \ P, As) with a program
3 2
built from modiÐed versions of two standard programs:
Gaussian 92/DFT18 for the quantum mechanics (QM) part
and MM3 (92) for the molecular mechanics19a (MM) part.
The QM part was always carried out for [Ir(C H )X(QH ) ]
4
4
3 2
at the computational level described in the previous para-
graph. For the MM part, the MM3(92) force Ðeld was used.19b
Van der Waals parameters for the iridium atom were taken
from the UFF force Ðeld.20 Parameters for bending contribu-
tions involving AswCwC bond angles and torsional con-
tributions involving As-C-C-R dihedral angles were taken
New J. Chem., 1998, Pages 1493È1498
1497

View Article Online
18 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. W. Wong, J. B. Foresman, M. A. Robb, M. Head-
Gordon, G. S. Replogle, R. Gomperts, J. L. Andres, K. Raghava-
chari, J. S. Binkley, C. Gonzalez, R. L. Martin, D. J. Fox, D. J.
Defrees, J. Baker, J. P. Stewart and J. A. Pople, Gaussian 92/DFT,
Gaussian, Inc., Pittsburgh, PA, 1993.
111, 8551; (c) J. H. Lii and N. L. Allinger, J. Am. Chem. Soc., 1989,
111, 8566; (d) J. H. Lii and N. L. Allinger, J. Am. Chem. Soc., 1989,
111, 8576.
20 A. K. Rappe C. J. Casewit, K. S. Colwell, W. A. Goddard III and
W. M. Ski†, J. Am. Chem. Soc., 1992, 114, 10024.
19 (a) N. L. Allinger, MM3(92), QCPE, Indiana University, 1992; (b)
N. L. Allinger, Y. H. Yuh and J. H. Lii, J. Am. Chem. Soc., 1989,
Received in Cambridge, UK, 25th June 1998; Paper 8/04840A
1498
New J. Chem., 1998, Pages 1493È1498